Attenuation map reconstruction from emission and transmission SPECT data

Attenuation correction is an essential step in single photon emission computed tomography (SPECT) reconstruction, and the required attenuation map can be determined from a transmission scan. The major drawback of this procedure is limited count statistics, due to time and dose considerations. It is well known that emission data also contains information about the attenuation distribution. In order to utilize this information, we have developed an attenuation map reconstruction method, which incorporates both emission and transmission data. The method is based on an existing transmission ML-EM algorithm, and the aim is to improve the photon statistics and reduce the noise in the final image. The method was evaluated using simulated data, and the results show that this new method can result in attenuation maps with a lower noise level and a more uniform noise distribution, as compared to using transmission data only.